MEPS 234:229-246 (2002)  -  doi:10.3354/meps234229

Importance of pre-recruitment life-history stages to population dynamics of the woolly sculpin Clinocottus analis

Jana L. D. Davis1,*, Lisa A. Levin2

1Scripps Institution of Oceanography, 9500 Gilman Drive, San Diego, California 92093-0218, USA
2Integrative Oceanography Division, Scripps Institution of Oceanography, 9500 Gilman Drive, San Diego, California 92093-0218, USA
*Present address: Smithsonian Environmental Research Center, 647 Contees Wharf Road, Edgewater, Maryland 21037, USA. E-mail:

ABSTRACT: The relative influence of pre- versus post-recruitment life-history events on population size has been the subject of much recent debate. In the marine realm, much work has focused on intertidal invertebrates and on tropical reef fishes, with mixed results. We addressed this problem for a temperate intertidal fish, Clinocottus analis. Our main goal was to determine which life-history stage was most responsible for temporal changes in population size from 1996 to 2000 at 2 sites in San Diego, California, both seasonally and during the 1997 to 1998 El Niño Southern Oscillation (ENSO) event. We approached the problem using cohort analysis and matrix population modeling. Recruitment pulses were evident in population size structure for up to a year, unobscured by post-recruitment mortality, which was not density-dependent. Recruitment was not correlated to spawning adult biomass of 3 mo earlier, suggesting that egg, larval, or early post-settlement processes during those 3 mo determined the magnitude of recruitment, and ultimately, population size. Stage-structured population projection matrices were constructed to compare population growth rates and sensitivities among seasons and between climate periods (El Niño and non-El Niño). Elasticity (prospective) and decomposition (retrospective) analyses of these matrices indicated that the vital rates to which population growth rate (λ) was theoretically most sensitive were not necessarily those responsible for observed temporal differences in λ. Although λ was most sensitive to juvenile growth and adult survivorship, fertility (which in this model included fecundity and egg, larval, and early post-settlement survivorship), in addition to juvenile growth, drove observed seasonal differences in λ. C. analis population size decreased during the 1997 to 1998 El Niño event due to a decrease in recruitment, a decrease in batch fecundity (hydrated eggs per female) and, at 1 site, changes in juvenile survivorship. Results of the study emphasize the power of early life-history events to structure C. analis populations on both seasonal and longer timescales.


KEY WORDS: Population dynamics · Population matrix models · Elasticity analysis · Decomposition analysis · Intertidal zone · Clinocottus analis


Full text in pdf format